Nanohelical Permeable Electronics with Strain-Insensitive Performance Authors: Zhenhua Yang, Zhehui Zhao, Chenning Li, Jiaye Li, Lei Cui, Yingtao Zhao, Chao-Peng Wang, Jian Wang, Wei Wang, Xin Zhang, Zhengmeng Yuan, Xiang Li, Sanchuan Zhao, Yuxuan Lin, Shiyu Liu, Banghao Wu, Zhaoqin Han, Jun Yin, Guanwen Liang, Yunmin Chen, Haoran Fu, Jian Zhu Published: 2026 ABSTRACT Stretchable electronics that are permeable to air and moisture are essential for emerging on-skin healthcare applications and human–machine symbiosis, yet a materials-agnostic strategy capable of delivering permeability together with performance-invariant stretchability across diverse electronic materials remains elusive. Here we report a universal and scalable approach for fabricating nanohelical electronic materials by transforming pre-stretched electrospun nanofibres into three-dimensional helical architectures via mismatch-strain-driven buckling. This geometry-enabled strategy accommodates a broad range of functional materials, including metals and organic/inorganic semiconductors, yielding large-area, lightweight and highly permeable nanohelical networks with low effective stiffness and pronounced strain insensitivity. Enabled by their spring-like deformation and intrinsic twist–bend coupling, the nanohelices redistribute strain uniformly and maintain stable electrical performance with negligible drift under large mechanical deformation, even at strains exceeding 50%. Finite-element analysis, together with a semi-analytical scaling model tailored to the half-coated nanofibre geometry, elucidates the underlying mechanics and relates the nanohelix morphology to fibre diameter, coating thickness and modulus contrast. We further demonstrate the versatility of this platform across multiple device modalities, including dry epidermal electrodes, stretchable electroluminescent displays, deformable interconnects, and strain-insensitive pressure, temperature and photodetectors, highlighting its potential for conformal and durable soft-electronic systems. Introduction Skin-conformal electronics represent a key pathway toward human-machine symbiosis1–3, enabling next-generation platforms for health monitoring and intelligent diagnostics4–6. The skin’s inherent deformability and dynamic perspiration pose a major challenge for wearables in maintaining long-term, stable conformity without compromising skin health. Therefore, incorporating air- and moisture-permeable functions into conformal electronics is crucial for ensuring reliable and skin-friendly interfaces7–9. Significant progress has been made in creating permeable substrates via thickness reduction10–14 or porous/mesh-like designs6,15–21 (Supplementary Fig. 1 and Table 1), leading to the realisation of permeable electrodes, interconnects, or hybrid integrated systems. Yet key challenges remain in developing permeable conductors and semiconductors with strain-insensitive electronic performance within the deformation regime (typically <50%) relevant to human body mechanics22,23, which are essential for intrinsically stretchable electronics. Soft organic materials24–27, such as conjugated polymers or conductive hydrogels, are promising candidates for stretchable and permeable electronics. Yet their intrinsically low carrier mobilities and conductivities restrict the breadth of their applications, and their electrical resistance is typically strain-sensitive due to strain-induced bond disruption or the separation of conductive domains28. In comparison, inorganic electronic materials29,30, which underpin modern information technologies, offer superior charge-ransport performance. However, they are inherently rigid and typically withstand only minimal strain due to strong covalent bonding. Although blending inorganic materials with elastomers can enhance stretchability, the insulating polymer matrix and incompatible interfaces often disrupt percolation pathways, resulting in pronounced strain-dependent resistance changes and limited composite processability31. An alternative strategy leverages geometry engineering, in which inorganic layers on elastomeric substrates are patterned into serpentine32,33 or kirigami34,35 two-dimensional (2D) architectures or transformed into wrinkled three-dimensional (3D) structures36–38 to reduce the effective strain during deformation. Nevertheless, these 2D structures are generally space-intensive, and 3D wrinkled architectures offer limited stretchability owing to stress localization within the inorganic layers during the “prestretch-release” process39,40. Additionally, their strain and conductance remain strongly coupled due to abrupt local stress redistribution during deformation. A helix represents a promising 3D architecture for achieving high stretchability, as it accommodates substantial and reversible elongations by redistributing axial strain into torsion and bending. Such mechanical strategies are widely exploited in nature, exemplified by DNA, plant tendrils and lotus-root fibrils41. In addition, substrate-unconstrained helices feature smoothly varying, uniform stress distributions, enabling small local strains and negligible variation in resistance during deformation42. Various approaches have been developed to fabricate helical electronic materials—for instance, mechanically guided assembly for creating helical interconnects43 and overtwisting44,45 for producing helical fibres. However, their reliance on complex procedures and stringent material combinations limits their scalability, particularly for large-area permeable electronics. Moreover, because helices intrinsically offer only uniaxial expansion, the resulting devices are typically restricted in their deformation modes unless multiaxial helical structures are introduced through substantially more sophisticated fabrication schemes. Here, we report a universal and scalable strategy for producing diverse helical electronic nanomaterials by transforming quasi-aligned or randomly oriented electrospun nanofibres through a buckling-enabled assembly process. Functional materials are thermally deposited onto pre-stretched thermoplastic polyurethane (TPU) nanofibres and, upon relaxation, spontaneously form stretchable metallic and semiconducting helices. A semi-analytical scaling model is established for the half-coated nanofibre geometry, which converts released prestrain into an effective natural curvature and predicts the helix diameter and pitch from the coupled bending–twisting relaxation of the rod. The resulting percolated and perforated networks provide conductive pathways with negligible resistance variations under omnidirectional stretching, while conformally interfacing with textured skin surfaces without impeding perspiration (Fig. 1a). These conductive helices function as permeable dry electrodes capable of high-fidelity physiological signal acquisition and as transparent electrodes for stretchable electroluminescent displays. The strain-insensitive nature of the helical architecture under reversible deformation is further demonstrated in permeable pressure, temperature and light sensors, all fabricated entirely from electrospun fibres and exhibiting high stretchability and robust performance (Fig. 1b). These conformal, permeable electronic devices demonstrate stable adhesion to the skin under both compressive and tensile mechanical deformation (Fig. 1c). Such a geometry-mediated design strategy fundamentally differs from previous prestretch/release approaches46,47 by transforming asymmetric nanofibre bilayers into nanohelical architectures capable of simultaneously achieving stretchability, permeability and electromechanical decoupling across diverse electronic materials.Fig. 1Concepts and fabrication of nanohelical permeable electronics with high stretchability.a Schematic diagram showing the excellent skin conformability of nanohelices for seamless integration with the textured skin surface. b Schematic of nanohelical permeable electronics for demonstration. c Photographic images showing the deformation of nanohelical films on human skin during stretching and compression. d Schematic diagram of the preparation of the nanohelical structures with various electronic materials (EM) on the surface. e Helices of alternating chirality connected by perversions along an individual nanofibre. f Scalable production of aligned nanofibres using high-speed rotating drum. g, Photographs showing large-area nanohelical networks with great transparency. h SEM image of large-area quasi-aligned nanohelices. Concepts and fabrication of nanohelical permeable electronics with high stretchability. a Schematic diagram showing the excellent skin conformability of nanohelices for seamless integration with the textured skin surface. b Schematic of nanohelical permeable electronics for demonstration. c Photographic images showing the deformation of nanohelical films on human skin during stretching and compression. d Schematic diagram of the preparation of the nanohelical structures with various electronic materials (EM) on the surface. e Helices of alternating chirality connected by perversions along an individual nanofibre. f Scalable production of aligned nanofibres using high-speed rotating drum. g, Photographs showing large-area nanohelical networks with great transparency. h SEM image of large-area quasi-aligned nanohelices. Results Scalable preparation of helical nanomaterials from electrospun nanofibres Helical structures were attained via thermally depositing nanoscale-thick materials onto pre-stretched electrospun TPU nanofibres with fixed ends, followed by gradual relaxation to their original length (Fig. 1d). Owing to the directional nature of thermal deposition, the functional materials are preferentially coated on one side of the TPU nanofibres, resulting in a crescent-shaped concave cross-section (Supplementary Fig. 2). The pronounced modulus mismatch between the deposited layer and the compliant TPU substrate then drives controlled buckling during relaxation, yielding helices of alternating chirality connected by perversions (kinks) along individual nanofibres (Fig. 1e). The deposited shell preserves its continuous morphology even after complete removal of the TPU nanofibres using dimethylformamide, confirming that the metal coating itself forms a continuous and mechanically stable framework rather than a collection of disconnected fragments (Supplementary Fig. 3). A rich variety of functional electronic materials—including metals (Ag, Au, Cu, Al, Cr, Ga) and semiconductors (Te, Copper(II) phthalocyanine or CuPc, pentacene and rubrene)—can be deposited onto the pre-strained fibres and reliably transformed into uniform helical architectures upon fibre relaxation (Fig. 1d). Energy Dispersive Analysis of X-rays confirms that these functional layers reside on the outer surface of the resulting helices (Supplementary Fig. 4). It should be noted that a continuous coating is essential for transforming nanofibres into nanohelices, because only a continuous shell can generate the interfacial stress mismatch required to drive the subsequent buckling-induced helical deformation (Supplementary Fig. 5). The nanohelices exhibit spring-like mechanical behaviour, accommodating both compressive and tensile deformation (Supplementary Fig. 6). This compliant behaviour enables nanohelical thin films to maintain conformal, stable adhesion to human skin during motion (Fig. 1c). This strategy is applicable to both aligned and randomly oriented nanofibre assemblies. Aligned nanofibres were collected and uniformly strained using a custom apparatus comprising two parallel, distance-tunable electrodes to ensure reproducible pre-stretching (Supplementary Fig. 7). Alternatively, random nanofibre networks were collected on an aluminium foil and subjected to omnidirectional pre-strain using a biaxial stretching platform48 (Supplementary Fig. 8). Furthermore, this nanohelical fabrication process was readily scaled by collecting aligned TPU nanofibres on a high-speed rotating drum (Fig. 1f and Supplementary Fig. 9). The drum speed dictates the residual strain stored in the fibres, ranging from 60% at 500 rpm to nearly 110% at 1000 rpm. Consequently, the fibres inherently possess sufficient prestrain and can be directly used for material deposition, forming helices upon release of the built-in stress without requiring intentional pre-stretching. Using this assembly strategy, we have successfully produced large-area nanohelical networks measuring 5 cm × 15 cm (Fig. 1g and Supplementary Fig. 10). SEM imaging further reveals that the quasi-aligned helical nanofibres form inter-fibre contacts after relaxation due to slight misalignment, which facilitates the development of percolated conductive networks (Fig. 1h). Mechanism of helical structure formation Helical diameter (Dh) and pitch (Ph) are key geometric descriptors of nanohelices and are governed by the interplay between the bending stiffness of the bilayer system and the release of stored elastic energy during relaxation. To elucidate these governing relationships, we systematically examined the influence of nanofibre diameter, Young’s modulus and deposited-layer thickness on these parameters under a fixed prestrain of 100%. We find that Dh is predominantly determined by the dimension of the TPU nanofiber. Specifically, Dh increases with the diameter of the TPU nanofibre, rising from 1.2 µm to 2.1 µm as the fibre diameter increases from 252 nm to 538 nm (Fig. 2a and Supplementary Fig. 11). Notably, Dh exhibits a remarkable insensitivity to the properties of the deposited material: it remains nearly invariant (~1.6 µm for a ~ 375 nm fibre) regardless of whether the coating is soft Te (E = 42 GPa) or stiff Cr (E = 201 GPa) (Fig. 2b), and shows negligible dependence on shell thickness (20–100 nm) (Fig. 2c). This behaviour confirms that the radial geometry is geometrically constrained by the TPU scaffold rather than the ultrathin coating.Fig. 2Experimental results and finite element analysis (FEA) of nanohelix formation.a Dependence of helical diameter (Dh) on nanofibre diameter, with the corresponding FEA results presented. Data points represent mean ± SD (n = 3), independent experiment 3 times. b Dependence of Dh on the type of deposition material, with the corresponding FEA results presented. Data points represent mean ± SD (n = 3), independent experiment 3 times. c Dependence of Dh on deposition layer thickness, with the corresponding FEA results presented. Data points represent mean ± SD (n = 3), independent experiment 3 times. d i) 3D model of ultrathin metal coating on a prestretched nanofibre; ii) 3D model of ultrathin metal coating on a prestretched TPU film. e, Strain contours of the nanohelical structure (i) and wrinkled structure (ii) for the 3D models in d) after relaxing. Experimental results and finite element analysis (FEA) of nanohelix formation. a Dependence of helical diameter (Dh) on nanofibre diameter, with the corresponding FEA results presented. Data points represent mean ± SD (n = 3), independent experiment 3 times. b Dependence of Dh on the type of deposition material, with the corresponding FEA results presented. Data points represent mean ± SD (n = 3), independent experiment 3 times. c Dependence of Dh on deposition layer thickness, with the corresponding FEA results presented. Data points represent mean ± SD (n = 3), independent experiment 3 times. d i) 3D model of ultrathin metal coating on a prestretched nanofibre; ii) 3D model of ultrathin metal coating on a prestretched TPU film. e, Strain contours of the nanohelical structure (i) and wrinkled structure (ii) for the 3D models in d) after relaxing. In comparison, Ph is intrinsically sensitive to the mechanical stiffness of the deposited layer. Both higher Young’s modulus and increased coating thickness lead to a marked expansion in Ph. For instance, under identical fibre dimensions, switching from Ag (E = 71 GPa) to stiffer Cu (E = 128 GPa) increases Ph from 3.6 µm to 4.9 µm (Supplementary Fig. 12). Similarly, the Ph of Ag helices increases from 2.3 µm to 3.3 µm as the shell thickness increases from 20 nm to 100 nm (Supplementary Fig. 13). Stiffer or thicker shells impose a higher energetic penalty for bending and torsion, thereby restricting the attainable curvature and driving the system toward relaxation through a longer pitch. These experimental observations are fully corroborated by finite element analysis (FEA), which captures the distinct sensitivities of Dh and Ph (Fig. 2a–c and Supplementary Fig. 13), validating the specific buckling mechanics of this core–shell architecture. To rationalise the observed helix geometry, a semi-analytical scaling model was developed in which the half-coated fibre is treated as an asymmetric bilayer rod (Supplementary Note 1). The released mismatch strain generates an effective natural curvature, which is further modified by torsion in a reduced Kirchhoff-rod description. Combining this energy minimisation with the geometry of a helical centreline gives1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\frac{{D}_{\mathrm h}}{{D}_{\mathrm f}}=\frac{1}{{\varepsilon }_{\mathrm r}{\Phi }_{\kappa }\left(\eta,\beta,\rho_{\mathrm G} \right)\left[1+{q}_{\mathrm g}^{2}{\Phi }_{\mathrm q}^{2}\left(\eta,\beta,\rho_{\mathrm G} \right)\right]}$$\end{document}DhDf=1εrΦκη,β,ρG1+qg2Φq2η,β,ρGand2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\frac{{P}_{\mathrm h}}{{D}_{\mathrm f}}=\frac{\pi {q}_{\mathrm g}{\Phi }_{\mathrm q}\left(\eta,\beta,\rho_{\mathrm G} \right)}{{\varepsilon }_{\mathrm r}{\Phi }_{\kappa }\left(\eta,\beta,\rho_{\mathrm G} \right)\left[1+{q}_{\mathrm g}^{2}{\Phi }_{\mathrm q}^{2}\left(\eta,\beta,\rho_{\mathrm G} \right)\right]}$$\end{document}PhDf=πqgΦqη,β,ρGεrΦκη,β,ρG1+qg2Φq2η,β,ρG Here, Df, tm, Gm, Gs, εpre denote the pristine nanofibre diameter, coating thickness, coating shear modulus, TPU core shear modulus, and prestrain, respectively; η = Em/Es, β=tm/Df, ρG = Gm/Gs, and εr=εpre/(1+εpre). The geometric factor qg is set by the axial projection constraint, while Φκ and Φq are calibrated from FEA. This semi-analytical model was validated by its good agreement with the FEA simulations and experimental results (Supplementary Fig. 14). The formation of this helical geometry is mechanically driven by the relaxation of mismatch strain within the asymmetric bilayer architecture. The directional deposition creates a crescent-shaped cross-section (Supplementary Fig. 2), where the asymmetric profile shifts the neutral axis away from the geometric centre of the fibre core. Upon relaxation, the compressive force exerted on the stiff coating generates an intrinsic bending moment due to this structural eccentricity. While this moment could, in principle, be accommodated through planar bending, the slender geometry renders such planar modes energetically unstable against out-of-plane deformation. Consequently, the system transitions into a coupled twist–bend mode49, dissipating the compressive stress through the formation of a 3D helix. This helical pathway is energetically favoured over pure in-plane bending, as it allows the stiff coating to relax via uniform curvature and torsion with minimal localized strain concentration (Supplementary Fig. 15), yielding the uniform nanohelices observed experimentally as the global equilibrium configuration50. The dynamic evolution of helical architectures in an Ag-coated (50 nm) pre-stretched TPU fibre clamped at both ends was further visualised by optical microscopy during relaxation (Supplementary Movie 1). As the tensile preload is released, multiple segments along the fibre began to rotate and wind about their longitudinal axis, giving rise to localized helical domains. Transitions between neighbouring helices of opposite handedness manifest as perversions, a motif characteristic of tendril-bearing climbers51. These perversions typically arise at geometric or material irregularities, such as local diameter enlargements or defect sites (Supplementary Fig. 16), which locally perturb the resistance to coupled bending–twisting deformation and redistribute deformation along the fibre, thereby selecting the locations of perversions between segments of opposite chirality as the overall length contracts (Supplementary Fig. 17 and Supplementary Note 2). FEA faithfully reproduces this sequence of twist–buckling events, capturing the spatiotemporal symmetry breaking that governs the helical morphology (Supplementary Movie 1). We further elucidated the uniqueness of this instability mode by contrasting the global helical buckling of nanofibres with the local surface wrinkling of bilayer films, a typical architecture in stretchable electronics52(Fig. 2d). The fundamental difference stems from the mechanical confinement imposed by the substrate. In planar bilayer systems, the massive substrate imposes a strong constraint on the ultrathin functional coating, prohibiting global twisting; consequently, the coating can only relax strain through short-wavelength out-of-plane wrinkling. FEA reveals that this is a geometry-governed transition: as the substrate width and thickness decrease to approach the coating dimensions, the dominant instability mode shifts from local surface wrinkling to torsional deformation, and ultimately to global helical buckling (Supplementary Fig. 18). This geometry-dependent mode transition can be summarised in a substrate-geometry-based regime map using Ω=min(W, d)/λw and χ = W/d, where W and d are the width and thickness of the substrate, and λw is the classical wrinkling wavelength (Supplementary Fig. 19 and Supplementary Note 3). This geometric distinction has profound implications for the local strain levels within the functional materials. For nanoscale metal films deposited on thick, pre-stretched elastomers, the relaxation process imposes severe in-plane compression, forcing the metal layer into high-curvature wrinkles that generate large peak strains. In contrast, in bilayer nanofibre systems where the core diameter is comparable to the shell thickness, the compliant substrate permits the entire structure to undergo global buckling. This allows the stiff coating to accommodate length changes through 3D geometric reconfiguration (uncoiling/coiling), thereby minimising local strain within the structure. Experiments and FEA collectively validate this mechanism: under 100% pre-strain, a 50 nm Ag layer in a wrinkled planar film experiences a maximum strain of up to 13%, leading to fracture (Fig. 2e and Supplementary Fig. 20); whereas in the helical nanofibre structure, the maximum strain remains below 5%, ensuring the integrity of the metal coating. In addition, the maximum strain in the Ag shell decreases with increasing applied tensile strain, as the imposed deformation is accommodated predominantly through reversible geometric deformation of the nanohelices rather than direct stretching of the metallic shell, thereby preventing excessive local strain accumulation and enabling repeated stretching without structural failure (Supplementary Fig. 21). Together, these results demonstrate that the helical architecture accommodates mechanical deformation through coupled twist–bend modes, thereby effectively decoupling electronic transport from mechanical strain at the nanoscale. Physical properties of nanohelical networks Nanofilms composed of nanohelical networks (NHNs) exhibit ultralow areal mass densities of 0.06 g cm−2 and are sufficiently lightweight to be lifted by a 10 µm-diameter copper wire (Fig. 3a). Owing to their highly porous architecture, these films are permeable to moisture and display water evaporation rates comparable to those of an open container across a range of temperatures (Fig. 3b). Ag-coated NHNs, whether aligned or randomly oriented, combine high optical transparency with supreme electrical conductivity. Both the optical transmittance and sheet resistance are governed by the network density, which can be precisely tuned by adjusting the electrospinning collection time (Fig. 3c and Supplementary Fig. 22). Aligned and random NHNs with 50-nm thick Ag exhibited optical transmittances of 81.5% and 80.5% at 550 nm, together with sheet resistances of 13.5 ± 1.6 Ω sq−1 and 52 ± 2.3 Ω sq−1, respectively—values comparable to those of previously reported metal-based transparent conductors53. It should be noted that an Ag thickness of only 20 nm is sufficient to generate well-defined nanohelices (Supplementary Fig. 13) and establish a continuous NHNs with a sheet resistance of 87 ± 5 Ω/sq. Increasing the Ag thickness further reduces the sheet resistance, while the sheet conductance (the reciprocal of the sheet resistance) scales linearly with Ag thickness, indicating continuous and uniform growth of the conductive coating (Supplementary Fig. 23). In addition, NHNs typically display markedly lower effective stiffness than uncoated TPU nanofibre networks, as only small forces are required to elongate the helical structures (Supplementary Fig. 24). This combination of ultralight weight, high permeability, optical transparency and mechanical compliance makes NHNs particularly attractive for permeable and conformal electronic-skin applications6,54.Fig. 3Physical properties of nanohelical networks (NHNs).a Photograph showing lightweight NHNs lifted by a copper wire with a diameter of 10 µm. b Comparison of water evaporation rate at different temperatures between uncovered and NHNs-covered containers. c Transmittance of aligned and random NHNs with different collection times. d Schematics showing the stretching directions for aligned and random nanofiber networks (NFNs) and NHNs. e Relative resistance versus strain for NHNs and NFNs. f Comparison of relative resistance changes of Ag-coated NHNs with those of previously reported stretchable conductors. g Cycling stability of Ag-coated random NHNs at a repeated strain of 50%. h Cycling stability of Ag-coated aligned NHNs at a repeated strain of 50%. i Relative resistance change of Ag–aNHNs under repeated pressing (~3 N). j Relative resistance change of TPU nanofibre membrane-protected Ag–aNHNs during rubbing against a rotating drum (~1 N). k Relative resistance change of Ag–aNHNs during simulated washing (400 r.p.m.). Physical properties of nanohelical networks (NHNs). a Photograph showing lightweight NHNs lifted by a copper wire with a diameter of 10 µm. b Comparison of water evaporation rate at different temperatures between uncovered and NHNs-covered containers. c Transmittance of aligned and random NHNs with different collection times. d Schematics showing the stretching directions for aligned and random nanofiber networks (NFNs) and NHNs. e Relative resistance versus strain for NHNs and NFNs. f Comparison of relative resistance changes of Ag-coated NHNs with those of previously reported stretchable conductors. g Cycling stability of Ag-coated random NHNs at a repeated strain of 50%. h Cycling stability of Ag-coated aligned NHNs at a repeated strain of 50%. i Relative resistance change of Ag–aNHNs under repeated pressing (~3 N). j Relative resistance change of TPU nanofibre membrane-protected Ag–aNHNs during rubbing against a rotating drum (~1 N). k Relative resistance change of Ag–aNHNs during simulated washing (400 r.p.m.). A distinctive feature of conductive NHNs is their exceptional resistance insensitivity under deformation, originating from the architecture-level strain accommodation enabled by reversible helix uncoiling55,56. To quantify this behaviour, we compared the electromechanical responses of NHNs with those of uncoiled nanofibre networks (NFNs), prepared by directly depositing Ag onto aligned or random TPU nanofibres. For aligned networks, electrical properties were measured parallel (∥) and perpendicular (⊥) to the fibre orientation, yielding four configurations: aNHNs∥, aNHNs⊥, aNFNs∥ and aNFNs⊥(Fig. 3d). Random networks exhibit isotropic electrical responses and are denoted as rNHNs and rNFNs (Fig. 3d and Supplementary Fig. 25). Aligned NHNs exhibited negligible resistance change (ΔR/R0 ≈ 1) up to strains of 98% (∥) and 89% (⊥), followed by a sharp increase beyond threshold strains of ~100% and ~200%, respectively (Fig. 3e). Strain accommodation occurs through helix uncoiling along the alignment and through network separation maintained by interhelical contacts in the transverse direction (Supplementary Figs. 26 and 27). The abrupt increase in resistance beyond the threshold strain is attributed to fracture of the Ag shells (Supplementary Fig. 28). Random NHNs similarly showed slow resistance evolution (ΔR/R0 ≈ 1 at 110%) before a threshold response, owing to multidirectional uncoiling of interconnected helices (Fig. 3e). In contrast, aligned NFNs displayed a rapid resistance increase (ΔR/R0 ≈ 50) at ~7% strain along the alignment due to Ag fracture during straight fibre deformation, while random NFNs reached ΔR/R0 ≈ 50 at ~15% strain through fibre sliding and coating failure (Fig. 3e). Benchmarking at 50% strain demonstrates the superior electromechanical stability of Ag-coated NHNs (ΔR/R0 = 0.035, 0.06, and 0.025 for aNHNs∥, aNHNs⊥, and rNHNs, respectively) compared with previously reported stretchable conductors, including liquid-metal and nanowire–elastomer composites38,53,57–63 (Fig. 3f and Supplementary Table 5). Moreover, the resistance variation remained nearly unchanged across different strain rates and sample dimensions, highlighting the robust electromechanical performance of Ag-coated NHNs (Supplementary Fig. 29). We further benchmarked the electromechanical performance of NHNs with that of previously reported conductive NFNs18,53,64–70 (Supplementary Table 6). For rigid metal coatings, NHNs consistently exhibit smaller resistance variations under tensile deformation. In conventional metal-coated NFNs, tensile strain is directly transferred to the conductive layer, resulting in crack formation and increased resistance. Although incorporating liquid metals into nanofibre networks can partially alleviate this limitation18, such approaches generally sacrifice the ultrathin and transparent characteristics of nanofibre networks while introducing potential risks associated with liquid-metal leakage. NHNs also demonstrated outstanding cyclic stability, with relative resistance changes of 0.06%, 0.02%, 0.001% per cycle for aNHNs∥, aNHNs⊥, and rNHNs, respectively over 200 cycles at 50% strain (Fig. 3g, h and Supplementary Fig. 30). Moreover, these electrodes maintained nearly identical ΔR/R0 and morphology during 10,000 loading cycles, further confirming their electromechanical robustness (Supplementary Fig. 31). By comparison, wrinkled Ag films on prestretched TPU exhibited resistance drift of ~1% per cycle at 10% strain, increasing to ~3% per cycle at 30%, owing to progressive crack accumulation (Supplementary Fig. 32). This resistance insensitivity is generalisable to other conductive NHNs, as demonstrated by Au-coated aNHNs (0.01% per cycle) and Te-coated aNHNs (0.3% per cycle), endowing the conductors with additional functionalities such as biocompatibility and photosensitivity (Supplementary Fig. 33). Moreover, owing to their freestanding and ultraconformal nature, NHNs can be readily transferred and robustly adhered to diverse flexible or elastomeric substrates, including polyethylene terephthalate (PET), Ecoflex, polydimethylsiloxane (PDMS), TPU nanofibre mats and VHB tape, yielding stretchable conductors with minimal resistance variation under strain and negligible drift during repeated deformation (Supplementary Figs. 34 and 35). The measured peel strengths of NHNs on compliant substrates, including Ecoflex and TPU nanofibre mats (Supplementary Fig. 36), are approximately 2 mN mm−1, confirming stable physical adhesion dominated by conformal contact and interfacial van der Waals interactions. The permeability of the NHNs can be preserved after transfer by perforating the otherwise non-breathable substrate (Supplementary Fig. 37). The practical durability of the NHNs was further evaluated under conditions relevant to daily use. Under repeated compression on human skin (3 N per cycle), the electrode exhibited only a 7% increase in resistance after 500 loading cycles (Fig. 3i). To improve abrasion resistance, an ultrathin TPU nanofibre layer was laminated onto the NHNs without compromising their air permeability or mechanical compliance. The resulting composite electrode showed only an 8% resistance increase after a cumulative rubbing distance exceeding 40 m (Fig. 3j). Stable skin adhesion was maintained with no noticeable morphological degradation after 24 h of continuous on-skin wear (Supplementary Fig. 38). In addition, the NHNs demonstrated excellent washing durability, exhibiting negligible resistance variation after 3 h of stirring in water and only an 11% increase after a subsequent 1 h ultrasonic cleaning (Fig. 3k and Supplementary Fig. 39). These results collectively demonstrate the excellent mechanical robustness and environmental stability of the NHNs for practical epidermal electronics. Nanohelical conductors for stretchable epidermal electrodes, interconnects, and displays Ag-coated NHNs are well suited for epidermal electrodes owing to their high conformability, permeability and electrical conductivity. Aligned NHNs were selected because they can be readily scaled using rotating-drum collectors. As shown in Fig. 4a, compared with commercially available hydrogel electrodes, Ag–aNHNs exhibit markedly lower skin–electrode contact impedance under both dry and perspiration conditions at frequencies below 1 kHz. This performance arises from the intimate conformal contact of Ag–aNHNs with the skin, whereby the deformable nanohelices adaptively conform to skin microstructures (Supplementary Fig. 40). Although individual nanohelices possess a 3D morphology, the NHNs behave mechanically as an ultralow-stiffness porous membrane. During attachment to skin, the low bending and torsional stiffness of the nanohelices allows individual conductive fibres to deform locally and accommodate the microscopic ridges and valleys of the skin with a low elastic energy cost (Fig. 1a). Once conformal contact is established, the interface is further stabilised by van der Waals interactions. Consequently, Ag–aNHNs maintain intimate skin contact even during perspiration and under water immersion (Supplementary Fig. 41). Consistent with this mechanism, 90° peel tests on human skin reveal comparable peel strengths of 2.2 mN mm−1 under both dry and wet conditions, comparable to those reported for representative dry epidermal electrodes71. Moreover, the skin–electrode impedance further decreases during perspiration owing to enhanced ionic conduction at the interface.Fig. 4Nanohelical conductors for stretchable and permeable electronics.a Skin–electrode contact impedance of Ag–aNHNs compared with a commercial hydrogel electrode. b Skin condition after prolonged wear using Ag–aNHNs and a hydrogel electrode, highlighting the improved skin compatibility of NHNs. c ECG and EMG signals recorded using Ag–aNHNs under dry and perspiration conditions. d Schematic illustration of a stretchable electroluminescent display incorporating patterned Ag–aNHNs electrodes. e SEM image of a patterned helical region on an individual nanofibre. f Optical image of the top patterned Ag–aNHNs electrode. g–i Stretchable electroluminescent display under biaxial deformation: g no applied strain; h vertical stretching; i, horizontal stretching. j Photographs of a stretchable LED circuit employing aNHN interconnects. k LED circuit under 100% tensile strain. l LED circuit conformally transferred onto the skin surface. m LED circuit on skin under compressive deformation. n LED circuit on skin under tensile deformation. Nanohelical conductors for stretchable and permeable electronics. a Skin–electrode contact impedance of Ag–aNHNs compared with a commercial hydrogel electrode. b Skin condition after prolonged wear using Ag–aNHNs and a hydrogel electrode, highlighting the improved skin compatibility of NHNs. c ECG and EMG signals recorded using Ag–aNHNs under dry and perspiration conditions. d Schematic illustration of a stretchable electroluminescent display incorporating patterned Ag–aNHNs electrodes. e SEM image of a patterned helical region on an individual nanofibre. f Optical image of the top patterned Ag–aNHNs electrode. g–i Stretchable electroluminescent display under biaxial deformation: g no applied strain; h vertical stretching; i, horizontal stretching. j Photographs of a stretchable LED circuit employing aNHN interconnects. k LED circuit under 100% tensile strain. l LED circuit conformally transferred onto the skin surface. m LED circuit on skin under compressive deformation. n LED circuit on skin under tensile deformation. Ag–aNHNs also offer excellent breathability and do not cause skin irritation during prolonged wear, in contrast to the erythema commonly induced by thick hydrogel electrodes (Fig. 4b). To complement the qualitative skin photographs, we performed a semi-quantitative analysis72 of skin redness using the red-to-green (R/G) intensity ratio extracted from RGB images (Supplementary Fig. 42). After 6 h of continuous wear, the R/G ratio of skin beneath commercial gel electrodes increased from 0.8 for untreated skin to 1.0, consistent with visible erythema. In contrast, the nanohelical electrodes maintained an R/G ratio comparable to that of untreated skin, indicating negligible skin redness after prolonged attachment. Electrocardiogram (ECG) and electromyogram (EMG) signals acquired using Ag–aNHNs are comparable to those obtained with hydrogel electrodes under dry conditions (Fig. 4c vs. Supplementary Fig. 43). Notably, Ag–aNHNs enable continuous, high-fidelity electrophysiological signal acquisition during perspiration, where commercial hydrogel electrodes typically fail owing to weakened interfacial adhesion and increased skin impedance73,74 (Supplementary Figs. 44, 45 and Fig. 4c). Under dry conditions, Ag–aNHNs achieve an ECG signal-to-noise ratio (SNR) of 33.1, which further increases to 34.7 during perspiration. In contrast, the SNR of commercial hydrogel electrodes decreases from 30.1 to 25.6 after sweat exposure, highlighting the excellent sweat tolerance and stable skin contact of the nanohelical electrodes. Moreover, Ag–aNHNs maintain nearly undistorted ECG and EMG recordings during voluntary arm movements or in the presence of vibration from a nearby motor, whereas commercial hydrogel electrodes exhibit pronounced motion artefacts. Stable ECG and EMG recordings are also achieved with the electrodes fully immersed in water, demonstrating their robustness under wet environments and their potential for aquatic wearable applications (Supplementary Fig. 46). The intrinsically non-invasive nature of Ag–aNHNs also allows seamless integration with daily wearable systems while supporting long-term, medical-grade physiological monitoring, thereby establishing a robust functional interface between the human body and smart electronic devices (Supplementary Movie 2). Ag–aNHNs can be readily patterned using stencil masks during thermal deposition (Fig. 4d, e). Feature sizes of 100 µm can be reduced to approximately 50 µm following helical formation, reflecting the geometrical contraction associated with the buckling process (Fig. 4e and Supplementary Fig. 47). These patterned nanohelical architectures enable diverse applications in flexible electronics, particularly for stretchable electroluminescent devices and highly deformable interconnect circuits. As illustrated in Fig. 4d, the electroluminescent display adopts a multilayer configuration comprising a top patterned Ag–aNHNs electrode (Fig. 4f), a middle ZnS/Ecoflex composite emissive layer, and a bottom Ag–aNHNs electrode. The device emits stable blue luminescence under an alternating voltage of 150 V at 500 Hz (Fig. 4g). Remarkably, the emission intensity remains almost unchanged under biaxial tensile strains of up to 100% (Fig. 4h, i and Supplementary Movie 3) or during dynamic hand movements (Supplementary Fig. 48). Quantitative luminance measurements show that approximately 95% of the initial luminance is retained at 100% strain, with a slight increase observed at strains below 30%. Moreover, the luminance decreases by only 2% after 1000 stretch–release cycles (Supplementary Fig. 49), demonstrating excellent electromechanical durability. This robust performance is attributed to the ability of the Ag–aNHNs electrodes to maintain stable electrical conductivity throughout the reversible deformation regime. In addition, a stretchable light-emitting diode (LED) circuit was fabricated by integrating patterned Ag–aNHNs with a commercial microchip (YX-1006) and discrete electronic components, including resistors, capacitors and LEDs (Fig. 4j and Supplementary Fig. 50). The resulting circuit maintains stable operation under tensile strains of up to 100%, underscoring the outstanding mechanical compliance and electrical stability of the NHNs (Fig. 4k and Supplementary Movie 4). To facilitate epidermal integration, we further developed a simple water-assisted transfer process that enables direct lamination of the permeable nanohelical electronics onto human skin with intimate conformal contact and robust adhesion (Fig. 4l). The resulting epidermal devices exhibit excellent electrical stability and operational reliability under various mechanical deformations associated with natural skin motion, including stretching and compression (Fig. 4m, n and Supplementary Movie 5). Strain-insensitive nanohelical pressure, temperature, and light sensors A stretchable and permeable pressure sensor based on capacitive transduction was fabricated by sandwiching a TPU/ionic liquid (IL) nanofibre mat between two Ag–aNHNs electrodes (Fig. 5a). TPU/IL nanofibres were employed instead of pristine TPU nanofibres to exploit electric double-layer (EDL) effects, thereby amplifying capacitance modulation under applied pressure (Supplementary Fig. 51). The device exhibits a linear pressure response over the range of 0–35 kPa with a sensitivity of 34.6 pF kPa−1 (Fig. 5b). Upon compression, the Ag–aNHNs electrodes undergo a reversible 3D–to–2D structural transition, markedly increasing the effective electrode–dielectric contact area and thus enhancing the capacitive response. The sensor also demonstrates excellent mechanical robustness, with no observable performance degradation under repeated loading, consistent with the exceptional resilience of Ag–aNHNs, which preserve their structural integrity after 100 compression cycles (Supplementary Fig. 52). By contrast, replacing Ag–aNHNs with Ag–aNFNs yields a substantially reduced capacitance response and poor cycling stability, owing to the absence of recoverable helical geometries in NFNs (Supplementary Fig. 53). Notably, the nanohelical pressure sensor exhibits pronounced strain insensitivity. Even under 50% tensile strain, the pressure sensitivity remains nearly unchanged at 34.2 pF kPa−1. Strain insensitivity, defined as the retention ratio of pressure sensitivity under applied strain75, reaches 0.99 at 50% strain, outperforming previously reported stretchable pressure sensors75–81 (Fig. 5c). Cyclic tensile testing at 50% strain further confirms the device robustness, with highly consistent response profiles and reproducible signal outputs over 100 cycles (Fig. 5 d). The sensor can also sustain 10,000 cycles of pressing with unvaried performance (Supplementary Fig. 54). Furthermore, patterned Ag–aNHNs were integrated into a stretchable 5 × 5 pressure-sensor array, in which each sensing pixel rapidly and accurately detects localized stimuli with minimal crosstalk (Supplementary Fig. 55).Fig. 5Strain-insensitive nanohelical sensors.a Schematic of the pressure/temperature sensor consisting of Ag-aNHNs electrodes and a TPU/ionic liquid composite nanofibre dielectric layer. b Capacitance responses under different applied pressures for sensors with identical architectures using NHNs or NFNs electrodes. c Benchmarking against previously reported stretchable pressure sensors in terms of maximum operational strain and strain insensitivity. PNC: Porous nanocomposite. CuINPs: CuI nanoparticles. PUD: Polyurethane dispersion. d Cyclic stability of unstretched and stretched pressure sensors. The pressure is 30 kPa. e Comparison in terms of the thermosensitivity with other reported flexible temperature sensors. f Relative capacitance responses of the temperature sensor without strain and at 50% strain. g Schematic of the nanohelical photodetector using Au–aNHNs electrodes and a Te–aNHNs channel before and after stretching. h Current–voltage characteristics under dark conditions and 520-nm illumination for unstretched and 50%-stretched device. i Photoswitching behaviour of the unstretched and stretched photodetector. j Temporal photoresponse of the photodetector. k Comparison with other reported TeNWs based photodetector in terms of response time. Strain-insensitive nanohelical sensors. a Schematic of the pressure/temperature sensor consisting of Ag-aNHNs electrodes and a TPU/ionic liquid composite nanofibre dielectric layer. b Capacitance responses under different applied pressures for sensors with identical architectures using NHNs or NFNs electrodes. c Benchmarking against previously reported stretchable pressure sensors in terms of maximum operational strain and strain insensitivity. PNC: Porous nanocomposite. CuINPs: CuI nanoparticles. PUD: Polyurethane dispersion. d Cyclic stability of unstretched and stretched pressure sensors. The pressure is 30 kPa. e Comparison in terms of the thermosensitivity with other reported flexible temperature sensors. f Relative capacitance responses of the temperature sensor without strain and at 50% strain. g Schematic of the nanohelical photodetector using Au–aNHNs electrodes and a Te–aNHNs channel before and after stretching. h Current–voltage characteristics under dark conditions and 520-nm illumination for unstretched and 50%-stretched device. i Photoswitching behaviour of the unstretched and stretched photodetector. j Temporal photoresponse of the photodetector. k Comparison with other reported TeNWs based photodetector in terms of response time. The same sensor architecture is also applicable for temperature sensing (Fig. 5a). Increasing temperature enhances ionic diffusion and migration within the TPU/IL mat while reducing the effective EDL thickness, resulting in an increase in capacitance. The temperature response was characterized over the physiologically relevant range of 30–42 °C, encompassing normal and febrile conditions, and exhibits a highly linear capacitance–temperature relationship (R2 = 0.99) with a sensitivity of 12% °C−1 (Supplementary Fig. 56), surpassing most reported flexible temperature sensors82–88(Fig. 5e). The temperature sensitivity remains essentially invariant under 50% tensile strain (Fig. 5f), highlighting the strain-independent sensing capability. The dynamic temperature response closely matches that recorded by a thermocouple, indicating fast response kinetics. Accordingly, the sensor enables real-time monitoring of transient thermal stimuli, such as heating induced by switching an 808 nm near-infra-red laser on and off, with rise and fall times of 110 ms and 800 ms, respectively (Supplementary Fig. 57 and Supplementary Movie 6). As a practical demonstration, the sensor was laminated onto the volar forearm for epidermal temperature monitoring, accurately resolving a skin temperature of 34 °C based on the calibrated linear response (Supplementary Fig. 58). The pressure and temperature sensors retained accurate and stable responses during body motion (Supplementary Fig. 59). A strain-insensitive nanohelical photodetector was realised using a two-terminal architecture comprising Au–aNHNs electrodes and a Te–aNHNs photosensitive channel (Fig. 5g and Supplementary Fig. 60). The device exhibits linear and symmetric current–voltage characteristics under both dark and illuminated conditions (520 nm), indicating near-ohmic Te/Au contacts. Only a minor reduction in current is observed under tensile strains up to 50%, demonstrating excellent mechanical tolerance (Fig. 5h, i). Upon illumination, the photocurrent increases markedly owing to photogenerated carrier accumulation within the Te nanohelices89,90. The photocurrent scales linearly with incident light intensity, with a linearity approaching 0.999. At an illumination power of 76.7 mW cm−2 and a bias of 0.1 V, the device delivers a strain-independent photocurrent of 2.32 μA, corresponding to a responsivity of 204 μA W−1. The relatively low responsivity is attributed to the high optical transparency of the nanohelical network, which allows most incident light to transmit through the device (Supplementary Fig. 61). Despite this, the photodetector exhibits fast temporal response, with rise and decay times of 130 ms and 260 ms, respectively—surpassing the performance of previously reported Te nanowire photodetectors89–95 (Fig. 5j, k). The permeable photodetectors retained accurate and stable photoresponses during on-hand ambient light monitoring, even under continuous hand movements (Supplementary Fig. 62). Discussion In conclusion, we present a universal and scalable strategy for constructing nanohelical electronic materials by converting pre-stretched electrospun nanofibres into 3D helical architectures through mismatch-strain–driven buckling. This approach supports a wide range of metals and semiconductors and enables large-area, lightweight and highly permeable nanohelical networks with low effective stiffness, uniform strain distribution and pronounced strain insensitivity. Owing to their spring-like deformation and twist–bend coupling, these nanohelices function as stretchable conductors with minimal resistance variation and robust adhesion, enabling stable operation in epidermal electrodes, electroluminescent displays, deformable interconnects and strain-insensitive pressure, temperature and photodetectors under large mechanical deformation. By leveraging geometry-mediated strain accommodation rather than material softening, this work establishes a general design paradigm for permeable, conformal and mechanically resilient electronic systems suitable for integration with dynamic soft and biological interfaces. Methods Preparation of nanohelices A TPU solution was prepared by dissolving 3 g of TPU (Grincoat) in 12 g of a dimethylformamide/tetrahydrofuran mixed solvent (1:1 by weight) and loaded into a 1 mL syringe for electrospinning. Electrospinning was performed using a 23 G needle at an applied voltage of 10 kV, a feeding rate of 0.5 mL h−1, and a collecting distance of 15 cm. Nanofibres were collected on parallel electrodes, aluminium foils, or rotating drum collectors to obtain aligned or random fibre networks. The as-spun nanofibre mats were pre-stretched to 100% strain, then thermally deposited with the selected electronic material. Upon release of the pre-strain, well-defined nanohelical structures formed spontaneously during elastic recovery. For nanohelix patterning, a PET stencil mask fabricated by laser cutting (xTool P2) was laminated onto the pre-stretched TPU nanofibre mat during thermal deposition. Upon release of the pre-strain, the patterned regions were relaxed and transformed into well-defined nanohelical networks. Preparation of wrinkled structures for comparison A TPU film was fabricated by spin-coating the TPU solution onto a glass substrate at 3000 rpm for 1 min, followed by baking at 100 °C for 1 h. The resulting film was exfoliated from the glass slide with the assistance of deionized water and cut into a rectangular shape (0.5 cm×3 cm). The TPU films were then pre-stretched to 100% strain, after which Ag was thermally deposited. Upon release of the pre-strain, well-defined wrinkling structures formed spontaneously. Fabrication of nanohelical sensors and electroluminescent display Te (50 nm) was first thermally deposited onto pre-stretched TPU nanofibres through a stencil mask to serve as the photosensitive layer. Subsequently, Au (50 nm) was deposited on both sides of the Te layer to form the source and drain electrodes. Upon release of the pre-strain, the device spontaneously transformed into a fully nanohelical photodetector architecture. TPU/ionic liquid (with a weight ratio of 15:1) nanofibre mats were fabricated by electrospinning using a 22 G needle at an applied voltage of 13 kV, a solution feed rate of 0.1 mL h−1, and a collection distance of 12 cm. The resulting mats were laminated between two Ag-coated nanohelical networks to form a sandwich-structured capacitor capable of sensing both pressure and temperature. ZnS electroluminescent powder (Shenzhen Yilai Technology Co., Ltd.) was dispersed into uncured Ecoflex and homogenised and degassed using a planetary mixer (AR-A00, THINKY). The composite was then cast into a film with a thickness of approximately 30 μm using a doctor-blade coating process. The resulting electroluminescent layer was laminated between a patterned and an unpatterned Ag-coated nanohelical network to form the display device. The electroluminescent display was driven by an AC voltage of 150 V at a frequency of 500 Hz. Finite element analysis Finite element analysis (FEA) was performed using the commercial software ABAQUS to simulate the helical buckling behaviour of circular-cross-section TPU wires with metallic coatings. The model consisted of a metallic layer and a TPU wire. The metallic coating was discretized with reduced-integration shell elements (S4R), while the TPU wire was meshed with hybrid solid elements (C3D8RH) to accurately represent its nearly incompressible hyperelasticity, mitigate volumetric locking, and ensure stability under large deformations. The TPU was modelled with a Mooney-Rivlin hyperelastic material model, with an elastic modulus of ETPU = 100 MPa and a Poisson’s ratio of μTPU = 0.49. Silver (Ag) was first chosen as the representative coating material, with EAg = 71 GPa and μAg = 0.37. To ensure computational accuracy, a refined mesh consisting of approximately 13,000 S4R elements for the metallic layer and 80,000 C3D8RH elements. Based on this setup, some models employed TPU substrates of different shapes to achieve wrinkling, bending, twisting, over-twisting and helical buckling modes, while the remaining models retained circular-cross-section TPU wires. To evaluate the effect of metallic type, additional models were built for copper (ECu = 128 GPa, μCu = 0.35), gold (EAu = 78 GPa, μAu = 0.40), chromium (ECr = 201 GPa, μCr = 0.30), and tellurium (ETe = 41.7 GPa, μTe = 0.33). To account for thickness and TPU wire size effects, Ag was used as the uniform coating material, while the metallic layer thickness and TPU wire diameter were systematically varied. Collection of ECG and EMG signals Electromyographic (EMG) signals were recorded using two electrodes positioned on the ventral forearm over the wrist flexor muscles, with an inter-electrode distance of 5 cm. A ground electrode was placed on the elbow. Electrocardiographic (ECG) signals were acquired by placing two electrodes symmetrically on the volar aspects of the right and left wrists, with a ground electrode positioned on the lateral malleolus. An ultrafine copper wire (diameter = 20 µm) was positioned beneath the Ag coated helical electrodes (1.5 cm×1.5 cm) to establish electrical connection with the data acquisition system. Data acquisition was performed using the BCIduino system (HANGYI BCI Network Co.), and was monitored in real-time using OpenBCI software sampled at a rate of 500 Hz. A 50 Hz notch filter was applied to mitigate power-line interference. Values of the signal-to-noise ratio (SNR) were calculated as per the following equation:3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{{\rm{SNR}}}}({{{\rm{dB}}}})=20\times {{\mathrm{lg}}}({{{{\rm{A}}}}}_{{{{\rm{s}}}}}/{{{{\rm{A}}}}}_{{{{\rm{n}}}}})$$\end{document}SNR(dB)=20×lg(As/An)where As is the average signal amplitude, An is the average noise amplitude. Experiments on human subjects All protocols involving human electrophysiological measurements underwent comprehensive ethical evaluation and received official clearance from Nankai University’s Institutional Review Board (reference number NKUIRB2023131). Each volunteer provided signed informed consent before commencing experimental procedures. Morphological and electrical characterizations The morphology of the nanohelices and devices was characterized by scanning electron microscopy (SEM; JSM-7800F, JEOL) and optical microscopy (Olympus BX-53). Cytocompatibility of the Au-coated nanohelical networks was evaluated using a Cell Counting Kit-8 (CCK-8) assay. Surface elemental distributions were analysed by energy-dispersive X-ray spectroscopy (EDS; EVO 18). Sheet resistance was measured using a four-probe method with a Keithley DAQ6510 digital multimeter. Resistance–strain characteristics were obtained using a linear translation stage coupled to a Keysight 34465 A digital multimeter at a stretching rate of 0.1 mm s−1, with resistance recorded in real time via a LabVIEW program. Copper tape and In–Ga liquid metal were used to ensure reliable electrical contact. All measurements were conducted at 25 ± 2 °C and a relative humidity of 55 ± 5% RH. Skin–electrode contact impedance was measured using an LCR metre (TH2838H, Changzhou Tonghui Electronics Co., Ltd.), with a pair of nanohelical electrodes (1.5 cm × 1.5 cm) placed 5 cm apart on the forearm of a human volunteer. Measurements were performed over a frequency range of 10 Hz to 1 MHz with an AC excitation voltage of 20 mV. Capacitance measurements for the pressure and temperature sensors were conducted using the same LCR metre. Temperature was controlled by a hotplate (BY-1010) and monitored in real time using a thermocouple (TT-K-30). Rapid thermal response was evaluated using an 808-nm laser modulated by a signal generator (Agilent 33521 A). Applied pressure was controlled and recorded using a motorised force tester (Mark-10) with a contact area of 15 mm × 8 mm. Photodetector performance under 520-nm laser illumination was characterized using a Keithley 4200A-SCS semiconductor parameter analyser. Statistics and reproducibility All experiments were repeated independently with similar results for at least 3 times. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Supplementary information Supplementary Information Description of Additional Supplementary Information Supplementary Movie 1 Supplementary Movie 2 Supplementary Movie 3 Supplementary Movie 4 Supplementary Movie 5 Supplementary Movie 6 Reporting Summary Transparent Peer Review file Supplementary Information Description of Additional Supplementary Information Supplementary Movie 1 Supplementary Movie 2 Supplementary Movie 3 Supplementary Movie 4 Supplementary Movie 5 Supplementary Movie 6 Reporting Summary Transparent Peer Review file Source data Source Data Source Data